Non-invasive patient respiration mechanical parameter monitoring method and system and respirator

Abstract

The invention belongs to the technical field of breathing machines, and particularly relates to a non-invasive patient breathing mechanical parameter monitoring method and system and a breathing machine. The method comprises the following steps: acquiring pressure and flow data at an airway inlet of a patient; integrating the flow data to obtain the gas volume within a set time length; constructing a motion equation about respiratory muscle pressure based on the first-order single-chamber respiratory system model; by applying physiological constraint conditions, respiratory muscle pressure is modeled into a sine function changing along with time and a preset waveform biased, so that the problem that parameter estimation is underdetermined due to unknown respiratory muscle pressure is solved; the preset waveform is substituted into the motion equation, fitting is carried out through the least square method, the respiratory system resistance, elasticity and respiratory effort of the patient are synchronously estimated, and non-invasive monitoring of the respiratory mechanical parameters of the patient is achieved. According to the invention, the problem that the prior art depends on passive respiration and invasively or cannot continuously monitor pain spots is solved.

Description

Technical Field

[0001] This invention belongs to the field of ventilator technology, and particularly relates to a non-invasive method, system and ventilator for monitoring patient respiratory mechanics parameters. Background Technology

[0002] Mechanical ventilation is a core intervention for patients with respiratory failure, but its improper use can easily lead to ventilator-related injuries. Accurate measurement and quantitative assessment of respiratory system mechanics parameters (hereinafter referred to as "respiratory mechanics parameters") are of crucial value for clinicians in managing mechanically ventilated patients.

[0003] Respiratory mechanics parameters include respiratory system resistance, elasticity, and respiratory effort. Respiratory system resistance (… rs This reflects the respiratory system's ability to resist airflow; respiratory system elasticity ( rs This reflects the tendency of a system to return to its original volume after being stretched. Clinically, the assessment of respiratory mechanics parameters often relies on the pressure at the airway inlet (…). P aw ) and flow ( Non-invasive measurement of ) is possible, but existing measurement methods all have significant limitations.

[0004] Inspiratory Hold Technique (IHT) is a clinically recognized method for monitoring respiratory mechanics. Its core principle is to rapidly block the patient's breathing circuit under constant inspiratory flow rate, simultaneously measuring the circuit flow and pressure before and after the blockage. This technique has two major advantages: first, it is non-invasive, requiring no invasive procedures; second, it is simple to operate, as most modern commercial ventilators are equipped with automated software that can automatically execute the blockage process and output results. rs , rs The result. But its limitations are equally prominent: 1) Interference with ventilation: Interruption procedures will disrupt the normal ventilation rhythm of the ventilator and make it impossible to continuously monitor respiratory mechanics; 2) Dependence on passive state: Measurement results are only reliable when the patient is completely passive (without voluntary breathing effort) throughout the entire inspiratory hold phase; 3) Inadequate fit: The respiratory mechanics of critically ill patients often change rapidly, and the inability to continuously monitor them can lead to parameter lag, affecting clinical decision-making.

[0005] To overcome the limitations of IHT, the Least Square Method (LSM) has become the mainstream alternative—its principle is to fit a mathematical model of the respiratory system with the pressure non-invasively measured at the patient's airway inlet. Paw ),flow( The data is fitted to estimate the resistance. rs ,elasticity rs .

[0006] Compared to IHT, the core advantage of LSM is that it does not interfere with normal mechanical ventilation, and parameter estimation can be achieved through two algorithms: Batch least squares algorithm: calls up complete data for the entire respiratory cycle and estimates values ​​for each breath. rs and rs This enables dynamic monitoring of respiratory mechanics; Recursive Least Squares (RLS) algorithm: Based on a formula design with a forgetting factor, it does not require a large amount of data storage and can track respiratory mechanics changes related to disease progression in real time, providing timely support for treatment plan adjustments.

[0007] However, LSM and IHT share a common limitation: they both require the patient to be in a completely passive state—if the patient is breathing spontaneously, the pressure generated by the respiratory muscles ( P mus This will become an undeniable driving force, causing the theoretical foundation of LSM to fail (unless...). P mus It can be used as a known input to the model.

[0008] To eliminate the interference of spontaneous breathing on the LSM, esophageal pressure can be introduced clinically. P es — Intrapleural pressure ( P pl Alternative metrics for LSM include airway pressure ( ), shifting the fitting object of LSM from airway pressure ( ) P aw Replace ) with transpulmonary pressure ( P aw - P es However, this improvement plan still has obvious shortcomings: 1) Limitations of assessment scope: It can only reflect the mechanical properties of the lungs and airways, and cannot take into account the contribution of the chest wall to elasticity; 2) Invasive procedure: It requires inserting a balloon catheter into the patient's esophagus, and requires a professional to perform catheter positioning and balloon inflation, making the procedure complex; 3) High technical threshold: It requires specialized equipment and is easily affected by operational errors and signal artifacts, resulting in low clinical applicability.

[0009] In summary, respiratory mechanics monitoring (especially non-invasive monitoring) during inspiratory activity in mechanically ventilated patients is not yet fully realized. Meanwhile, in critical care medicine, the application of some assisted ventilation modes, such as pressure support ventilation (PSV), is becoming increasingly widespread. While these modes can promote respiratory muscle activation, facilitate weaning from mechanical ventilation, and thus improve prognosis and reduce hospitalization costs, they place higher demands on the real-time performance and accuracy of respiratory mechanics monitoring. Therefore, developing non-invasive and accurate respiratory mechanics monitoring technologies adapted to spontaneous breathing remains one of the core directions of current clinical research and urgently needs to be addressed. Summary of the Invention

[0010] The purpose of this invention is to overcome the defects of the prior art and to propose a non-invasive method for monitoring patient respiratory mechanics parameters. This invention also discloses a non-invasive patient respiratory mechanics parameter monitoring system and a ventilator.

[0011] In view of this, the present invention proposes a non-invasive method for monitoring patient respiratory mechanics parameters, comprising: Step 1: Obtain pressure and flow data at the patient's airway inlet; Step 2: Integrate the flow rate data to obtain the gas volume within the set time period; Step 3: Construct the motion equations for respiratory muscle pressure based on a first-order single-chamber respiratory system model; Step 4: By applying physiological constraints, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform to solve the problem of underdetermined parameter estimation caused by unknown respiratory muscle pressure. Step 5: Substitute the preset waveform into the equation of motion and fit it using the least squares method to simultaneously estimate the patient's respiratory system resistance, elasticity, and respiratory effort, thereby achieving non-invasive monitoring of the patient's respiratory mechanics parameters.

[0012] As an improvement to the above method, the method further includes, before step 1: setting a monitoring mode for each respiratory cycle or a real-time monitoring mode.

[0013] As an improvement to the above method, the equation of motion for respiratory muscle pressure in step 3 is as follows:

[0014] in, This refers to the pressure at the airway inlet. For traffic data; It is the gas volume within a set time period. It is a constant term. The patient's respiratory muscle pressure cannot be measured directly. For respiratory system resistance, It is the reciprocal of the respiratory system's elasticity, or compliance.

[0015] As an improvement to the above method, the physiological constraints in step 4 include: The signal curve of respiratory muscle pressure is regular and does not show irregular changes during a single breath; During the initiation phase of spontaneous breathing, the pressure of the respiratory muscles decreases monotonically, and when the respiratory muscles enter a relaxed state, the pressure of the respiratory muscles increases monotonically. In passive exhalation, the respiratory muscle pressure remains at zero throughout the entire exhalation process.

[0016] As an improvement to the above method, in step 4, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform. for:

[0017] in, For the first k Each sampling time, N It is the total number of sampling moments during a single inhalation. Angular velocity, This represents the amplitude of the sine function. Indicates bias.

[0018] As an improvement to the above method, the coefficients of the sine function and bias The angular velocity was obtained by least squares fitting. Obtain it using one of the following methods: Set to a fixed value; Calculated based on the recorded inspiratory duration; Initialize or adjust the respiratory system resistance and elasticity values ​​estimated from the previous respiratory cycle.

[0019] As an improvement to the above method, step 5 includes: When a respiratory cycle-by-breath monitoring mode is adopted, the problem of estimating respiratory mechanics parameters is formulated as a constrained optimization problem with a cost function. for:

[0020] in, The superscript T indicates transpose. Indicates the time of the Nth sampling. This represents the pressure at the airway inlet at the Nth sampling time. This represents the flow rate data at the Nth sampling time; This represents the gas volume over the duration of N samplings. The cost function is optimized using the batch least squares method. To obtain respiratory system resistance Respiratory system elasticity Bias of the sine function Sum of coefficients The optimal estimate: .

[0021] As an improvement to the above method, step 5 includes: When using real-time monitoring, respiratory resistance is obtained through a recursive least squares method with a forgetting factor. Respiratory system elasticity Bias of the sine function ,coefficient of k Update the estimated value continuously: :

[0022]

[0023] in, for k The parameter estimation vector updated at each time step. The superscript T indicates transpose. for k The airway inlet pressure obtained from continuous sampling for k Time measurement vector, ; express k Traffic data obtained from real-time sampling; express k Gas volume within the sampling time period; for k Gain matrix at time step; The forgetting factor, which is between 0 and 1. , They are respectively k time, k- The observation covariance matrix at time 1.

[0024] Secondly, the present invention provides a non-invasive patient respiratory mechanics parameter monitoring system, the device comprising: A pressure sensor is used to acquire pressure data at the patient's airway inlet; A flow sensor is used to acquire flow data through the patient's airway; and The data processing module is used to perform the following operations: A motion equation for respiratory muscle pressure was constructed based on a first-order single-chamber respiratory system model. By applying physiological constraints, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform to solve the problem of underdetermined parameter estimation caused by unknown respiratory muscle pressure. By substituting the preset waveform into the equation of motion and fitting it using the least squares method, the respiratory system resistance, elasticity, and respiratory effort of the monitored patient are simultaneously estimated, thus achieving non-invasive monitoring of the patient's respiratory mechanics parameters.

[0025] On the other hand, the present invention provides a ventilator including a non-invasive patient respiratory mechanics monitoring system as described above.

[0026] Compared with the prior art, the advantages of the present invention are: The method of the present invention enables accurate monitoring of respiratory system resistance, elasticity and respiratory effort in patients with active breathing (especially in assisted ventilation modes such as pressure support ventilation) without interfering with normal ventilation of the ventilator and without the need for invasive esophageal manometry. It solves the pain points of existing technologies that rely on passive breathing, are invasive or cannot be continuously monitored.

[0027] This invention uses a first-order single-chamber mathematical model of the respiratory system to abstract the respiratory system into a structure of "elastic chamber + resistance pathway". Based on the first-order single-chamber model, physiological constraints are added and the assumption of "sine function plus bias" of respiratory effort is introduced to solve the problem of parameter underdetermined estimation caused by respiratory effort during spontaneous breathing.

[0028] This invention provides two specific monitoring implementation methods for different patients: Respiratory cycle monitoring: This method detects the start of inspiration (patient-triggered) and the transition from inspiration to expiration (patient switching), collecting airway pressure and flow data within the cycle and integrating them to obtain the volume. Angular velocity parameters are selected using various strategies, and then solved using batch least squares. This method requires storing all sampled data for a single respiratory cycle, resulting in a large memory footprint. It is suitable for situations where the patient's respiratory status is relatively stable and parameters change slowly.

[0029] Real-time monitoring: Employing a recursive least squares method with a forgetting factor, after initializing the parameter estimation vector and observation covariance matrix, the parameter estimation vector, gain matrix, and observation covariance matrix are updated according to the recursive formula at each sampling time, outputting the parameters in real time. This method does not require storing large amounts of historical data (only the parameters from the previous time step are needed, resulting in low memory usage) and is suitable for situations where respiratory mechanics parameters change rapidly, such as in critically ill patients. Attached Figure Description

[0030] Figure 1This is a schematic diagram illustrating the application of the non-invasive patient respiratory mechanics parameter monitoring method of the present invention; Figure 2 This is a flowchart of monitoring the patient's respiratory mechanics parameters on a respiratory cycle basis; Figure 3 It is a flowchart for real-time monitoring of a patient's respiratory mechanics parameters during an inspiratory-expiratory cycle; Figure 4 This is a schematic diagram of the patient breathing effort model proposed in this invention. Detailed Implementation

[0031] This invention proposes a novel method for non-invasive monitoring of respiratory mechanics parameters in patients with spontaneous breathing, without interfering with normal ventilator operation and without requiring esophageal manometry. The method measures pressure and flow at the patient's airway inlet and, based on a first-order single-chamber model, monitors respiratory system resistance, elasticity, and respiratory effort. The novelty of this method lies in considering the presence of spontaneous breathing, introducing a sinusoidal assumption of respiratory effort, and resolving the "parameter estimation underdeterminacy problem caused by spontaneous breathing," thereby... rs and rs The estimate is reliable even with respiratory effort.

[0032] The non-invasive method for monitoring patient respiratory mechanics parameters proposed in this invention relies on a medical ventilator. The core components of this device include: a ventilator body for providing ventilation support to the patient; and a pressure sensor deployed at the ventilator's Y-connector (for measuring the pressure at the airway inlet). P aw ) and flow sensor (used to measure the flow rate in and out of the patient's airway) ); and a respiratory system monitoring module. This monitoring module has a built-in microprocessor, and its core function is to estimate multiple respiratory parameters of the patient using the least squares method based on the sinusoidal assumption, specifically including: (i) respiratory system resilience or compliance ( rs or (ii) respiratory system resistance ( rs ), and (iii) respiratory muscle pressure ( P mus ( t )).

[0033] The technical solution of this invention estimates respiratory mechanical parameters based on the respiratory system motion equation. Using a first-order single-chamber respiratory system model, the respiratory system motion equation can be written in the following form: (1) in, The pressure measured at the Y-connector is used to replace the pressure at the patient's mouth. The flow rate entering the patient's respiratory system is measured by a flow sensor; It refers to the volume of gas delivered by the ventilator to the patient's respiratory system, measured by the flow rate. The result is obtained by time integration; It is a constant term used to account for airway pressure at the end of exhalation (used in the balance equation, it has no special significance), and will be incorporated into the following discussion. In the middle; This is the equivalent pressure exerted by the patient's respiratory muscles on the respiratory system, which cannot be directly measured. All of these are constants to be estimated. For respiratory system resistance, It is the reciprocal of respiratory system elasticity (compliance).

[0034] If the patient (In cases where spontaneous breathing is absent or suppressed by drugs or other means), by collecting pressure and flow data and combining them with the motion equation of equation (1), the equation can be solved by solving the simultaneous equations to obtain the result. Estimates of three constants. If the patient If the value is not constantly zero (indicating spontaneous breathing), the above estimation problem is an underdetermined problem, meaning it cannot be estimated using the same method. and There can be infinitely many estimates that satisfy equation (1) over a period of time. Simply put, assume... and They are respectively and The estimated value, given the same set of measurements ,make Satisfying equation (1) yields one solution to the above estimation problem; however, there are many other solutions that lack practical significance. In fact, we can assume... For any value, we can always find a value using the following formula. It satisfies the equality constraint of equation (1), that is (2) To simplify the mathematical expression, in equation (1), we let ,because The breathing rate remains constant throughout the entire breathing process. Therefore, the estimation problem can be formulated as a constrained optimization problem with a cost function, which is: (3) in, Representing the kEach sampling time is used because data is typically sampled discretely by the acquisition device. N This is the total number of sampling moments during a single inhalation. Assume the sampling interval is... If the initial time is zero, then .

[0035] This invention effectively overcomes the underdeterminacy of mathematical problems by imposing constraints on the unknowns to be estimated. Specifically, it is based on the following physiological principles: 1) The pressure signal curve exerted by the respiratory muscles has an inherent regularity and will not show irregular changes during a single breath; 2) During the initiation phase of spontaneous breathing, the pressure exerted by the respiratory muscles shows a monotonically decreasing trend, while when the respiratory muscles enter a relaxed state, the pressure shows a monotonically increasing trend. 3) During passive exhalation, the pressure exerted by the respiratory muscles remains zero throughout the entire exhalation process.

[0036] Based on the above physiological principles, this invention proposes to... Assuming it takes the form of "sine function with bias", such as... Figure 4 As shown, its specific expression is as follows: (4) Substituting equation (4) into equation (3) yields the following: (5) Rewrite the above equation in the form of a matrix equation. (6) in, By optimizing the cost function using the batch least squares method, we can obtain... The optimal estimate, i.e. (7) The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and embodiments.

[0037] Example 1 Embodiment 1 of the present invention proposes a non-invasive method for monitoring patient respiratory mechanics parameters, comprising the following steps: Step 1: Obtain pressure and flow data at the patient's airway inlet; Step 2: Integrate the flow rate data to obtain the gas volume within the set time period; Step 3: Construct the motion equations for respiratory muscle pressure based on a first-order single-chamber respiratory system model; Step 4: By applying physiological constraints, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform to solve the problem of underdetermined parameter estimation caused by unknown respiratory muscle pressure. Step 5: Substitute the preset waveform into the equation of motion and fit it using the batch least squares method to simultaneously estimate the patient's respiratory system resistance, elasticity, and respiratory effort, thereby achieving non-invasive, non-real-time monitoring of the patient's respiratory mechanics parameters.

[0038] Figure 1 The ventilator shown is a dual-circuit ventilator with proximal sensors and is for illustrative purposes only. It should be noted that the non-invasive method for monitoring patient respiratory mechanics parameters proposed in this paper is highly compatible and unrestricted, and can be used with almost all types of ventilators. Specifically, it includes four categories: first, by tubing design, covering single-circuit and dual-circuit ventilators; second, by power components, including ventilators with proportional valves and ventilators with fans; third, by patient connection method, including invasive connection types (such as tracheostomy or endotracheal tube connection) and non-invasive connection types (such as mask connection); fourth, by sensor configuration, including... Figure 1 The ventilator shown includes a proximal sensor with pressure and flow measurement functions, as well as ventilators that do not have such a proximal sensor and rely solely on sensors within the ventilator unit.

[0039] refer to Figure 1 Patients need to be monitored using various physiological parameter sensors. Specifically, Figure 1 The document showcases two core types of sensors: one is an airway pressure sensor, used to measure pressure at the patient's connection points. The first measurement point is typically located at the Y-connector; the second is a flow sensor, used to measure the flow rate of gas flowing into or out of the patient. The measurement point is usually set at the Y-connector. In terms of installation location, the airway pressure sensor and flow sensor can be integrated into the Y-connector, placed separately in the tubing, or directly integrated into the ventilator body.

[0040] The ventilator data acquisition module typically acquires the outputs of the airway pressure sensor and flow sensor at a fixed frequency. The optimal acquisition frequency is... Then the sampling interval .

[0041] This embodiment uses respiratory cycle monitoring, and the specific procedure is as follows: Figure 2 As shown.

[0042] When a patient trigger is detected, record the moment of inhalation onset. When a patient switch is detected, record the time from inspiratory to expiratory breathing. ,but .collection arrive airway pressure over a period of time and traffic data For traffic data Numerical integration is performed to obtain the gas volume delivered to the respiratory system. .

[0043] Angular velocity in equation (4) The parameter can be preferably set to a fixed value, or selected according to a certain strategy. One such strategy is to select the parameter based on the duration of inhalation. (8) In addition, angular velocity The parameters can also be obtained from monitoring the previous respiratory cycle. The value is determined, that is (9) in, A number between 3 and 5 can be chosen, preferably. .

[0044] Sure , and Then, calculate Value, get Vector sum Array values. To solve problem (7), according to the least squares principle, we know that (10) Calculate and output the results using the above formula to obtain the respiratory system resistance. Respiratory system elasticity Bias of the sine function Sum of coefficients The optimal estimate: Parameter monitoring is performed to complete one inhalation and exhalation cycle.

[0045] Example 2 Embodiment 2 of the present invention proposes a non-invasive method for monitoring patient respiratory mechanics parameters, comprising the following steps: Step 1: Obtain pressure and flow data at the patient's airway inlet; Step 2: Integrate the flow rate data to obtain the gas volume within the set time period; Step 3: Construct the motion equations for respiratory muscle pressure based on a first-order single-chamber respiratory system model; Step 4: By applying physiological constraints, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform to solve the problem of underdetermined parameter estimation caused by unknown respiratory muscle pressure. Step 5: Substitute the preset waveform into the equation of motion and fit it using the recursive least squares method with a forgetting factor. Simultaneously estimate the patient's respiratory system resistance, elasticity, and respiratory effort to achieve real-time monitoring of non-invasive patient respiratory mechanics parameters.

[0046] This embodiment uses real-time monitoring, and the flowchart is as follows: Figure 3 As shown.

[0047] Because breath-by-breath monitoring requires recording airway pressure, flow, and volume data from the start of inspiration to the transition from inspiration to expiration, it occupies a significant amount of memory. To save memory, a recursive least squares method with a forgetting factor can be used to optimize the monitoring process, thereby enabling real-time and continuous monitoring of respiratory mechanics parameters.

[0048] The recursive formula for least squares with a forgetting factor is as follows: (11) (12) (13) The specific meanings of each symbol during real-time monitoring of respiratory mechanics parameters are as follows: Representing the k Each sampling time; for k The parameter estimation vector updated at time step 1. ; for k The airway pressure data obtained from sampling at different times, i.e. ; for k The time-measured vector, i.e. ; for k Gain matrix at time step; The forgetting factor, ranging from 0 to 1, is typically set between 0.98 and 0.995, with 0.98 being the preferred value. ; for k The observation covariance matrix at time t.

[0049] The specific process is as follows: When a patient trigger is detected, the initial value of the estimated vector is given. Initial values ​​of the observed covariance matrix Estimating the initial value of the vector. Typically, it is assigned zero, or a value with a high probability of occurrence; in one embodiment, the first inhalation / exhalation cycle is preferred. The units are respectively , , and ; thereafter, each breathing cycle The initial values ​​are all the same as the final values ​​at the end of the previous period; the initial values ​​of the covariance matrix are... Usually, the larger number is taken, that is... , For a sufficiently large constant, For a diagonal identity matrix, the preferred choice is... .

[0050] At each subsequent sampling time Calculate and update according to formulas (11)(12)(13) and and output in real time value.

[0051] When a patient switch is detected, the recursive process is stopped and recorded. The final value is used as the initial value for the next period.

[0052] Example 3 Embodiment 3 of the present invention provides a non-invasive patient respiratory mechanics parameter monitoring system for performing the method of Embodiment 1 or 2, the system comprising: A pressure sensor is used to acquire pressure data at the patient's airway inlet; A flow sensor is used to acquire flow data through the patient's airway; The data processing module is used to perform the following operations: A motion equation for respiratory muscle pressure was constructed based on a first-order single-chamber respiratory system model. By applying physiological constraints, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform, thus solving the problem of underdetermined parameter estimation caused by unknown respiratory muscle pressure. By substituting the preset waveform into the equation of motion and fitting it using the least squares method, the respiratory system resistance, elasticity, and respiratory effort of the monitored patient are simultaneously estimated, thus achieving non-invasive monitoring of the patient's respiratory mechanics parameters.

[0053] It is worth noting that in the embodiments of the above system, the modules included are divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional module are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0054] Example 4 Embodiment 4 of the present invention provides a ventilator comprising the system of Embodiment 3. The type of ventilator is not limited. Specifically, it includes four categories: first, classified by tubing design, encompassing single-tubing and dual-tubing ventilators; second, classified by power components, including ventilators with proportional valves and ventilators with fans; third, classified by patient connection method, including invasive connection types (such as those via tracheostomy or endotracheal tube connection) and non-invasive connection types (such as those via mask connection); fourth, classified by sensor configuration, including... Figure 1 The diagram shows ventilators equipped with proximal sensors that measure pressure and flow, as well as ventilators that do not have such proximal sensors and rely solely on sensors within the ventilator unit. Table 1 shows the meanings of technical terms and English abbreviations.

[0055] Table 1. Meanings of proper nouns and English abbreviations

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that modifications or equivalent substitutions to the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A non-invasive method for monitoring patient respiratory mechanics parameters, comprising: Step 1: Obtain pressure and flow data at the patient's airway inlet; Step 2: Integrate the flow rate data to obtain the gas volume within the set time period; Step 3: Construct the motion equations for respiratory muscle pressure based on a first-order single-chamber respiratory system model; Step 4: By applying physiological constraints, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform to solve the problem of underdetermined parameter estimation caused by unknown respiratory muscle pressure. Step 5: Substitute the preset waveform into the equation of motion and fit it using the least squares method to simultaneously estimate the patient's respiratory system resistance, elasticity, and respiratory effort, thereby achieving non-invasive monitoring of the patient's respiratory mechanics parameters.

2. The non-invasive method for monitoring patient respiratory mechanics parameters according to claim 1, characterized in that, The method also includes, prior to step 1, setting either a respiratory cycle monitoring mode or a real-time monitoring mode.

3. The non-invasive method for monitoring patient respiratory mechanics parameters according to claim 2, characterized in that, The equation of motion for respiratory muscle pressure in step 3 is as follows: in, This refers to the pressure at the airway inlet. For traffic data; It is the gas volume within a set time period. It is a constant term. The patient's respiratory muscle pressure cannot be measured directly. For respiratory system resistance, It is the reciprocal of the respiratory system's elasticity, or compliance.

4. The non-invasive method for monitoring patient respiratory mechanics parameters according to claim 1, characterized in that, In step 4, the physiological constraints include: The signal curve of respiratory muscle pressure is regular and does not show irregular changes during a single breath; During the initiation phase of spontaneous breathing, the pressure of the respiratory muscles decreases monotonically, and when the respiratory muscles enter a relaxed state, the pressure of the respiratory muscles increases monotonically. In passive exhalation, the respiratory muscle pressure remains at zero throughout the entire exhalation process.

5. The non-invasive method for monitoring patient respiratory mechanics parameters according to claim 1, characterized in that, In step 4, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform. for: in, For the first k Each sampling time, N It is the total number of sampling moments during a single inhalation. Angular velocity, This represents the amplitude of the sine function. Indicates bias.

6. The non-invasive method for monitoring patient respiratory mechanics parameters according to claim 3, characterized in that, The coefficients of the sine function and bias The angular velocity was obtained by least squares fitting. Obtain it using one of the following methods: Set to a fixed value; Calculated based on the recorded inspiratory duration; Initialize or adjust the respiratory system resistance and elasticity values ​​estimated from the previous respiratory cycle.

7. The non-invasive method for monitoring patient respiratory mechanics parameters according to claim 3, characterized in that, Step 5 includes: When a respiratory cycle-by-breath monitoring mode is adopted, the problem of estimating respiratory mechanics parameters is formulated as a constrained optimization problem with a cost function. for: The superscript T indicates transpose. Indicates the time of the Nth sampling. This represents the pressure at the airway inlet at the Nth sampling time. This represents the flow rate data at the Nth sampling time; This represents the gas volume over the duration of N samplings. The cost function is optimized using the batch least squares method. To obtain respiratory system resistance Respiratory system elasticity Bias of the sine function Sum of coefficients The optimal estimate: .

8. The non-invasive method for monitoring patient respiratory mechanics parameters according to claim 3, characterized in that, Step 5 includes: When using real-time monitoring, respiratory resistance is obtained through a recursive least squares method with a forgetting factor. Respiratory system elasticity Bias of the sine function ,coefficient of k Update the estimated value continuously: : in, for k The parameter estimation vector updated at each time step. The superscript T indicates transpose. for k The airway inlet pressure obtained from continuous sampling for k Time measurement vector, ; express k Traffic data obtained from real-time sampling; express k Gas volume within the sampling time period; for k Gain matrix at time step; The forgetting factor, which is between 0 and 1. , They are respectively k time, k- The observation covariance matrix at time 1.

9. A non-invasive patient respiratory mechanics parameter monitoring system, characterized in that, The system includes: A pressure sensor is used to acquire pressure data at the patient's airway inlet; A flow sensor is used to acquire flow data through the patient's airway; and The data processing module is used to perform the following operations: A motion equation for respiratory muscle pressure was constructed based on a first-order single-chamber respiratory system model. By applying physiological constraints, the respiratory muscle pressure is modeled as a time-varying sine function with a biased preset waveform to solve the problem of underdetermined parameter estimation caused by unknown respiratory muscle pressure. By substituting the preset waveform into the equation of motion and fitting it using the least squares method, the respiratory system resistance, elasticity, and respiratory effort of the monitored patient are simultaneously estimated, thus achieving non-invasive monitoring of the patient's respiratory mechanics parameters.

10. A ventilator, characterized in that, Including non-invasive patient respiratory mechanics monitoring systems as described in claim 9.

Images